Concept Map Assessment Through Structure Classification

TL;DR

The paper addresses the challenge of assessing concept map quality by automatically classifying maps into Spoke, Chain, or Network structures based on extractable features from maps represented as directed graphs. It deploys feature engineering on 317 maps, evaluates five multiclass classifiers with 5-fold cross-validation, and finds that a Decision Tree achieves 86% accuracy, with key predictive features including the standard deviation and mean of edges per node, as well as higher-order edge-topology metrics. The study demonstrates the potential for real-time, automated concept map assessment to aid educators and learners, while acknowledging manual labeling as a limitation and proposing deeper structural analysis in future work. Overall, the approach provides a scalable pathway to integrate structure-aware feedback into concept-map-based learning environments.

Abstract

Due to their versatility, concept maps are used in various educational settings and serve as tools that enable educators to comprehend students' knowledge construction. An essential component for analyzing a concept map is its structure, which can be categorized into three distinct types: spoke, network, and chain. Understanding the predominant structure in a map offers insights into the student's depth of comprehension of the subject. Therefore, this study examined 317 distinct concept map structures, classifying them into one of the three types, and used statistical and descriptive information from the maps to train multiclass classification models. As a result, we achieved an 86\% accuracy in classification using a Decision Tree. This promising outcome can be employed in concept map assessment systems to provide real-time feedback to the student.

Figures (2)

• Figure 1: Examples of concept maps in three structures: A) spoke; B) chain; C) network. Source: Kinchin et al. (2005) KinchinHay2005
• Figure 2: Examples of spoke, chain, and network structures found in the maps